The twelve Jyotirliṅgas are the most completed pilgrimage circuit in Śaiva India. People spend years on it, and finishing all twelve is the kind of thing that gets mentioned.
The source makes an observation about those people that is worth sitting with.
Every Jyotirliṅga has a Bhairava dedicated to it. And many who have completed the circuit, if you ask them whether they saw the Bhairava at any of the twelve — they don't have the information at all.1
Not that they went and were turned away. They were standing in the kṣetra, at the site, and the thing was there, and they did not register it.
That is the page's actual subject. The claim about the overlay matters less than the fact that a whole layer of a famous circuit can be walked through by devoted people and not be seen.
The source gives the map in a run, and it arrives unevenly — some positions clear, some mangled, some skipped.
Somnath → Kapāla.2 Mallikārjuna → Siddha.3 Mahākāleśvara → Kāla directly.4 Omkāreśvara → named as the fourth svarūpa.5 Bhīmāśaṅkar → Bhīma, described as the governor of every state and a Kāla Bhairava himself.6 Tryambakeśvar → Vikrānta.7 Rāmeśvaram → Ānanda.8 One → Śarabha.9 Kāśī Viśvanātha → Kāla himself, as Mahākāla, kṣetrapāla of the entire Kāśī khaṇḍa.10
⚑ That is nine positions, not twelve. The remaining three are not given in the range read. The source moves on saying "when time comes we will talk about this particular svarūpa," and does not return.11 No page in this build should present this as a complete table. The Pass-1 inventory described PB as giving "roughly eight of twelve clearly," which is close and slightly pessimistic.12
The structurally interesting part is not the nine that are named. It is the two that are named as hidden.
Of the twelve, the source says, two are not disclosed — they are Pātāla. One is Vikrānta at Tryambakeśvar. The other is Mahākāleśvara.13
At Mahākāleśvara, Kāla himself is described as being in the group of Pātāla Bhairava.14 So the position is doubly occupied: the site's Bhairava is Kāla, and Kāla there is held as Pātāla.
Which produces a neat and slightly suspicious result. The circuit has twelve stations; ten are teachable; two belong to the form that has no public mantra, no yantra, and is transmitted orally only.15
A pilgrimage route with two undisclosable stops. You can complete the circuit and, on this account, still not have received two of its twelve.
⚑ Worth naming the structure without assuming design. A scheme that reserves part of itself as unavailable is exactly what a transmission-based tradition needs in order to remain necessary after the public itinerary is finished. It is also what you would expect if the material genuinely is restricted. The overlay does not let you distinguish these.
One concrete detail anchors the whole overlay, and it is the sort of thing that is hard to invent.
At Ujjain, in the Kāla mandira, there is a Pātāla Bhairava. And the offering there mirrors the famous one: where liquor is offered to Kāla Bhairava, at this position a silver bowl is offered by the pujārīs — and the source says it is the same shape of offering as is made to Kāla in the Kāśī kṣetra, because he is Kāla.16
Look at what that does. The identification is not being argued theologically. It is being read off ritual practice: the offering made here matches the offering made there, therefore the deity here is the deity there.
That is a real method and it is one scholars of religion use — cult practice is often more conservative than doctrine, and two sites making the same odd offering are more likely to share a history than two sites making the same claim.
It is also the only place in this episode where the overlay is supported by something other than assertion. The other eleven positions are stated. This one has a bowl.
⚑ And an honest limit: liquor offerings to Kāla Bhairava at Ujjain are famous and widely reported; the silver bowl at the Pātāla position is the source's own report and is not corroborated here.
Step back from the individual pairings and ask what kind of claim this is.
The Jyotirliṅgas are mainstream Śaiva territory — the most public, most institutional, most visited network in the tradition. Nothing about them is esoteric. They belong to everyone.
Laying twelve Bhairavas over them says: this network was never only what you thought it was. Each of these famous sites has a second occupant, and the second occupant is from the tantric register.
That is an annexation, and it is achieved without contradicting anything. No Jyotirliṅga is denied. No mainstream claim is refuted. A layer is simply asserted to have been present all along, and the evidence for its presence is that you did not notice it — which is the same structure the corpus uses for Dhūmāvatī's consort and for Bagalāmukhī's.17
⚑ The move is legible and that does not make it false. Guardian deities at major shrines are genuinely widespread in Indian practice, and kṣetrapāla figures at Śaiva sites are well attested. What this corpus adds is the systematic mapping — twelve to twelve, each named — and the systematic version is exactly the part that would be constructed if any of it were.
One of the twelve is given as Śarabha, and following that name outward turns up the sharpest contradiction in the episode.
The source attaches Śarabha to a hilltop Bhairava temple at Daśārāma in Andhra, sourced to the Skanda Purāṇa, and identifies the form there as Śarabhāvatāra — Śiva's own.22
Then the standard story: Narasiṃha's fury could not be contained, and Śarabha — a composite of animals, Śiva in that shape — came and controlled the situation. Thousandfold more powerful.23
And then, in the very next breath, the counter-version. But in the Devī Purāṇa it is spoken a different way. There Bhairavī emerged and controlled both Śarabha and Narasiṃha — because both were creating havoc out of their tāmasic guṇa. And some hold that Pratyaṅgirā was part of it, and that Pratyaṅgirā took control of Śarabha and Narasiṃha together.24
So within ninety seconds the same passage says Śarabha resolved the crisis and that Śarabha was part of the crisis, requiring a goddess to resolve him.
⚑ Preserved, not resolved. The source presents both without ranking them — "again, in Devī Purāṇa it is spoken into a different way" — and moves on to Khaṇḍobā.
Which is worth noticing precisely because it is the corpus at its best. It does not smooth the variant away or argue for one. It reports that the Śaiva telling and the Śākta telling disagree about who ended the fight, and leaves the disagreement standing — which is exactly what this vault's own rules ask for and what most of the rest of this material does not do.
Think of a place you know completely. A building you worked in for years, a route you walked daily, a town you grew up in.
Now get someone who knows it in a different register to walk it with you — a builder, a birder, a local historian, someone who lived there in a different decade.
What happens is reliable and slightly disturbing. They will point at things that have been in your visual field thousands of times and that you have never once seen. Not hidden things. A blocked-up doorway. A datestone. A species of tree.
You did not miss them through inattention. You missed them because nothing in your framing had a slot for them, and perception fills slots.
That is what the pilgrims did at the twelve. And the practice this page suggests is not to go looking for hidden layers — that produces a person who finds them everywhere.
It is narrower. Once, in a place you know well, deliberately ask: what would someone with a different training see here that I cannot? Then go and get them, and let them show you, and notice how thoroughly the thing was already visible.
The point is not the doorway. It is calibrating how much of what you look at every day does not reach you — because the next time somebody claims there is a layer you have been walking past, you will have a personal measure of how likely that is.
One video, December 2025, auto-generated captions, an unnamed teacher whose authority is his guru's blessing and his own experience.18
The overlay passage is among the worst-transcribed in the episode — Jyotirliṅga arrives as jot linga, joti linga, jotilinga and jar linga; several site and svarūpa names are lost entirely. Three of the twelve positions are not recoverable and are not reconstructed.19
No text is cited for the overlay. The Pātāla positions at Tryambakeśvar and Mahākāleśvara are asserted.
What's strong. The silver-bowl detail, because it grounds an identification in a practice rather than a claim, and because ritual detail is conservative. And the observation about pilgrims — that a whole layer of a famous circuit goes unregistered by people who completed it — is a genuinely testable claim about what people report, and would be easy to check.
Tension, preserved. The overlay is presented as recovering something that was always there, and its supporting evidence is that nobody noticed it. That is the corpus's standard shape and it is unfalsifiable in the same way each time.
Tension, second. Nine of twelve positions are given; the source promises the rest and does not return. Any use of this material as a complete table would be supplying three positions the corpus never stated.
Open question. If two of the twelve stations are undisclosable, what is the status of a completed pilgrimage? Either the circuit is completable without them — in which case they are not really stations — or nobody has completed it, which is a strong claim about the most-completed route in Śaiva India. The corpus does not choose.
The vault already holds the twelve Jyotirliṅgas from a Purāṇic ingest, where they are what they have always been: twelve sites, a canonical list, a geography of light.20 That account has no second layer and does not need one. The liṅgas are the subject.
Come here and the same twelve are a substrate. They are not what is being taught; they are what the teaching is being hung on. The Jyotirliṅga is the address and the Bhairava is the tenant.
That difference is not a disagreement about facts, and it is worth being precise about what it is instead. The Purāṇic account is a list of places. This is a list of guardians, indexed by place. Both can be true simultaneously, and a pilgrim could hold both without strain.
But they generate opposite instructions. The Purāṇic circuit is complete when you have visited twelve sites. This one is complete when you have met twelve beings — and since two of them are undisclosable, it cannot be completed by visiting. Which means the same journey, under the second description, is one nobody finishes, and the incompleteness is structural rather than a failure of effort.
A tradition that overlays guardians onto a completable route has converted it into an uncompletable one, and that conversion is doing work whether or not anyone intended it.
Psychology — the framing that decides what reaches you. The vault holds Langer's premature cognitive commitments: an early framing, adopted before evaluation, that governs what later perception admits — and the reason it is so hard to catch is that it operates upstream of judgement rather than downstream.21
The pilgrims are that mechanism, cleanly isolated.
They were not sceptical about Bhairavas at Jyotirliṅgas. Scepticism would require the category to have arrived at all. They had a complete and correct framing — this is a Jyotirliṅga circuit, twelve sites, here is what one does — and that framing had no slot for a second occupant. The shrines were in their visual field. Nothing reached them.
The insight neither produces alone: an esoteric layer and a nonexistent layer produce identical reports from the uninitiated, which is why "nobody notices it" can never be evidence for either. The psychology explains precisely why the non-noticing is worthless as data — a correct framing suppresses exactly as much as a defective one, and the pilgrims' failure to see would be equally complete whether the Bhairavas are there or not.
Which yields the discipline the corpus needs and lacks. A claim of the form this has always been here and you were not equipped to see it is not supported by the fact that nobody saw it. It requires something the framing could not have produced: a bowl, a staircase, a practice that persists without a reason. The overlay has one of those, at Ujjain, and eleven assertions.
Sharpest implication. Overlaying guardians onto the twelve Jyotirliṅgas converts the most completable pilgrimage in Śaiva India into one that cannot be completed — because two of its stations are, by the same doctrine, undisclosable — and nobody performing the circuit is told this.
Generative questions.
Footnotes 22–24 belong to a section added after the list was set; numbered out of reading order deliberately, since renumbering in place risks repointing an existing citation.